avgfp · Sequence CNN

reference_avgfp_cnn_contiguousf1_20261104-153920

Metrics

Set	mse	pearsonr	spearmanr	r2	above_wt_n	above_wt_spearmanr	budget_to_max	hit_rate_at_10	mean_at_10	max_at_10	captured_at_10	hit_rate_at_50	mean_at_50	max_at_50	precision_at_50	captured_at_50	hit_rate_at_100	mean_at_100	max_at_100	precision_at_100	captured_at_100	hit_rate_at_1pct	mean_at_1pct	max_at_1pct	precision_at_1pct	captured_at_1pct
train	0.2188	0.9063	0.7674	0.6801	1898.0000	0.2076	2512.0000	0.7000	3.7506	3.8826	0.6322	0.6000	3.7551	4.0212	0.0400	0.6896	0.6000	3.7346	4.0314	0.0500	0.6984	0.5818	3.7324	4.0314	0.0455	0.7007
validation	0.5382	-0.5401	-0.3550	-0.0783	142.0000	0.0569	1013.0000	0.0000	1.4108	1.7044	0.0000	-3.2010	0.0000	1.7628	3.5713	0.0000	0.0000	-3.0379	0.0200	2.2084	3.7670	0.0100	0.0000	-2.3494	0.0000	1.4250	1.7044	0.0000	0.0000	-3.2058
test	1.3245	-0.2475	-0.2241	-0.3175	60.0000	0.0774	291.0000	0.0000	2.2005	3.6624	0.0000	-0.5978	0.0200	2.2224	3.7520	0.0200	0.0000	-0.6425	0.0400	2.4582	3.8994	0.1100	0.0000	-0.4137	0.0000	2.4170	3.6624	0.0000	0.0000	-0.3820

Loss per epoch

training validation

Config

name: reference_avgfp_cnn_contiguousf1
task: dms
data:
  dataset: avgfp
  encoding: one_hot,aa_index
  split_dir: null
  split_scheme: contiguous
  split_fold: 1
model:
  name: cnn
  conv:
  - - 128
    - 3
  - - 128
    - 3
  - - 128
    - 3
  - - 128
    - 3
  - - 128
    - 3
  hidden:
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 64
  learning_rate: 0.001
  patience: 15
  min_delta: 1.0e-05
  seed: 7